Beyond the Pareto Efficient Frontier: Constraint Active Search for Multiobjective Experimental Design
Book • 2023
This work introduces constraint active search as a new approach to multiobjective design problems.
It emphasizes finding diverse samples that satisfy minimum performance thresholds rather than solely focusing on the Pareto efficient frontier.
The authors propose an algorithm called Expected Coverage Improvement (ECI) to efficiently explore regions of interest in the parameter space.
It emphasizes finding diverse samples that satisfy minimum performance thresholds rather than solely focusing on the Pareto efficient frontier.
The authors propose an algorithm called Expected Coverage Improvement (ECI) to efficiently explore regions of interest in the parameter space.
Mentioned by
Mentioned in 0 episodes
Mentioned by 
Gustavo Malkomes
Constraint Active Search for Human-in-the-Loop Optimization with Gustavo Malkomes - #505
